Judging by manufacturer announcements, car and airline executives have suddenly discovered the Internet. They want to get you online, no matter where you are or what you're doing. Everybody, from Boeing to Volvo, is promising access to E-mail and the Internet in planes and automobiles, and it looks likely to be a train wreck.

The most recent announcements have an air of "1999 all over again" about them: They look set to repeat the expensive mistakes by companies (such as OmniSky) that tried to persuade consumers to pay for mobile information access. Fortunately for everybody else in the mobile economy, this is a great opportunity to watch and learn from the highly funded, technology-led initiatives of the likes of Ford, Boeing, and General Motors.

Despite disappointing results to date, business-to-consumer mobile technologies can be successful, as long as they're designed for the user's needs, desires, and constraints, rather than the generic mobilization of the Internet. There has been growing interest, spearheaded by OnStar, in using telematics in vehicles. The services range from helpful (in-car navigation) to useful (remotely unlocking car doors if you locked yourself out) to potentially lifesaving: if the airbag deploys in a monitored car, somebody will call you on the built-in telephone and, if necessary, alert emergency services. However, in the last few months, car manufacturers seem to have lost sight of what makes current telematics systems appealing. Instead, they're loudly announcing plans to add services such as E-mail, Internet browsing, and even E-commerce transactions. Not to be left out, technology companies such as Sun Microsystems and Oracle (in its Wingcast partnership with Ford) also revealed intentions to be players in this market. Meanwhile, Boeing is pressing ahead with its plans for in-flight Internet access, although all but one of its airline partners in this project (Lufthansa) have dropped out. Most recently, Accenture and Microsoft revealed a partnership to bring .Net to the car dashboard, stating that "the future of auto computing is one where your home and work PCs are synched up with your car PC. You will be able to surf the Net, pull stock quotes, and get E-mail and instant messages via voice recognition." Stock Answers
What this latest round of announcements misses is that the telematics services that have been popular to date--and the services that, according to surveys, consumers remain willing to pay for--are ones that are specifically related to the task at hand. In the case of drivers, that means systems related to safety, security, navigation, and perhaps entertainment--not generic, purposeless Internet or E-mail access. On the other hand, this proposal does highlight one extremely useful rule of thumb: Any technology that's promoted with the promise that a primary benefit is the ability to track or trade stocks is almost certainly doomed. That's a sure red flag that the backers haven't thought of a single genuinely useful role for the technology. This isn't just an exercise in criticism, though. These proposals throw into relief the kinds of mobile services that are likely to succeed. Here's the key criteria: Offer the right content. How many people outside of Wall Street and Silicon Valley actually need to know stock prices, on a constant basis, to the extent that they would pay to receive them (as opposed to casually viewing them in a desktop portal)? This fixation on stock prices comes from projecting the needs or desires of those building or backing the service onto the potential buyers of the service--a classic mistake of the design process. Choose the right channel. Alongside stocks, these services inevitably propose news, weather, and sports. Together, they're known as the "Four Horsemen of the WAPocalypse," whose appearance portends the doom of a wireless consumer service. While the information may be of genuine interest to a large number of people, it's readily available via the mundane and conventional car radio, usually with far greater richness than a text-to-voice rendition of a written news story. It's hard to get people to pay to receive news, sports, or weather just a little sooner. Use the right technology. What is the advantage of receiving instant messages by voice recognition over simply receiving a phone call? Far more useful would be server-side services that determine your location and route communications accordingly--perhaps instant message if you are at your desk, phone if you are in your car, and voice mail or E-mail if you are unavailable. Leverage what exists already. Anybody with a pressing need for constant access to E-mail probably wants a RIM BlackBerry or other wirelessly enabled PDA, so they can get messages anywhere--not just in the car. The last thing they want is another E-mail address, another mail forwarder to manage, and another monthly subscription to pay. It would make more sense to be able to dock the existing device in a cradle in the car (or even have it connect itself to the car's environment wirelessly using Bluetooth) to use the car's systems, such as driving the PDA using voice recognition. Designing For The Users. Remember Them?
To create compelling mobile services, designers need to replace a technology-centric approach (and an obsession with mobile stock trading) with a user-centric approach. The first step in doing so is to identify the different types of users and their roles. When you think about drivers' primary responsibilities, and study drivers to see how they work, it's not hard to realize the value of the hazard warning system that's often integrated in radar detectors. Now imagine that this system were tied into the in-car navigation system, which automatically revised your route when alerted to an accident ahead. Go one step further and imagine that the system checks your calendar, sees that you're going to be late for your flight as a result of the delay, and automatically rebooks you on the next flight while sending an instant message to the person who was supposed to meet you at the other end. None of this is beyond the realm of current technology. As usual, poor design is more often caused by a lack of imagination than a lack of technology. We'll probably see a year or two of disappointingly generic "in-car information systems" whose value will mostly be to serve as design lessons for the rest of us, as well as knocking technology bugs out of complex, integrated wireless systems. After that, I firmly expect the car manufacturers to get back to what they--in normal times--excel at: identifying, understanding, and meeting the needs of the car user.